Automatic estimation of stacking fault density in SiC specimens imaged by transmission electron microscopy

At this moment the SiC wide band-gap semiconductor is one of the most commonly used for high temperature, high frequency and high power electronic devices and various light emitting devices. With the stacking faults representing one of the most important native defects of SiC, the stacking fault density represents a key characteristic in regard to the suitability of a particular SiC crystal or SiC growth method to be employed in the fabrication process of a SiC-based device. In this paper we show that the stacking fault density can be automatically calculated from TEM images of SiC crystals by using appropriate image processing and computer vision techniques.